Schizophrenia affects more than 2 million Americans and leads to profound disability. Medications can reduce positive symptoms but do not specifically target functioning. Functional outcome shows stronger relationships with cognitive impairments and social skills deficits than with positive symptoms. Psychosocial interventions combined with pharmacotherapy can improve cognitive and social functioning, but the appropriate cognitive targets and optimal combination of intervention techniques are unclear. Well-controlled trials of psychosocial interventions that amplify pharmacotherapy by alleviating cognitive impairments and improving community functioning are needed. We developed, manualized and piloted cognitive-behavioral social skills training (CBSST), which targets cognitive and metacognitive impairments, social skills deficits, and thoughts that interfere with functioning in schizophrenia patients. The proposed project is a randomized controlled trial comparing CBSST with Supportive Contact (SC). Patients with schizophrenia (N=236) will be randomized to one of the 2 conditions, treated for 1 year and followed longitudinally for 1 year after treatment. A multi-dimensional evaluation of outcomes will include social functioning (primary outcome), neuropsychological functioning, cognitive insight, psychiatric symptoms, and health services utilization. The short-term objectives are to examine the effectiveness of CBSST over SC and potential predictors and mechanisms of change. Key questions are: (1) Does adding CBSST to standard care improve rehabilitation outcomes relative SC in schizophrenia patients? (2) Does severity of neurocognitive impairment at baseline or improvement in neurocognitive impairment in treatment differentially predict social functioning outcome? (3) Does improvement in cognitive insight mediate improvement in psychotic symptoms in CBSST? If CBSST continues to produce incremental effectiveness, future larger effectiveness and dissemination trials would be justified.